Giter VIP home page Giter VIP logo

lungmap-pipeline's Introduction

lungmap-pipeline

Complete analysis pipeline for the segmentation & classification of structures / cells for LungMap images

How to Use

This repository is meant to be a demonstration of how the microscopy-analysis-pipeline can be used to integrate with lungmap data to effect instance segmentation and classification of anatomical structures. As such, we provide several python scripts which demonstrate how to use the resources together. But first, to run the scripts, a little setup...

Environment

We assume a Python 3 environment with dependencies listed in the requirements.txt file.

pip install -r requirements.txt

Data

For the python scripts to run properly, we assume that the lungmap data has been downloaded and symlinked as data into this repositories parent folder.

cd ~
git clone https://github.com/duke-lungmap-team/lungmap-image-data.git
cd /path/to/lungmap-pipeline
ln -s ~/lungmap-image-data data

Running example/*.py

Check the top of each script before running. Some scripts assume that other scripts have been run first. This is done to stash model objects or other metadata needed to produce further output.

lungmap-pipeline's People

Contributors

benneely avatar dependabot[bot] avatar whitews avatar

Watchers

 avatar  avatar

Forkers

benneely

lungmap-pipeline's Issues

examples/run_pipeline.py fails

In debugging, it looks like this is failing in cv2_extras.utils.find_border_by_mask, specifically, for some contours the function scipy.optimize.curve_fit is being fed a range, which doesn't have a shape attribute. Let me know if you can replicate.

Trouble reading ontology .owl file

Using ontospy==1.9.8.2, I'm getting a fatal error in gui.utils when trying to instantiate an ontospy.Ontospy object. The error is:

Fatal error parsing graph <../resources/lung_ontology.owl>
(using RDF serializations: ['xml'])

Windows 10 Bug: Generate Candidate Contours Triggers new LungMAP Region Generator screen

On Windows 10, upon selecting default settings and then selecting the image *_001.tif and then pressing Find Regions, the pipeline starts to find contours and we see the following in the console:

Pre-processing test image...
90 color candidates found, kernel: (15, 15)

at which point a new blank LungMAP Region Generator Image screen populates. This is occurring in lungmap_pipeline.Application.run_segmentation. Not sure if threading from find_regions is the culprit. Let me know if you have ideas about tracking this down.

Extra Utilities Question

@whitews - In order to generate evaluations, I need to stash a few utility functions. Considering creating an examples/utils folder for this purpose. Please verify this is acceptable?

Question about examples

I'm adding instructions to the README to tell users how to get their own copy of the data for running the example pipelines. However, currently the lungmap-image-data repository only contains 20x images. How will we handle the 60X examples?

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.